Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Electrocardiography wikipedia , lookup
Heart failure wikipedia , lookup
Myocardial infarction wikipedia , lookup
Antihypertensive drug wikipedia , lookup
Remote ischemic conditioning wikipedia , lookup
Cardiac surgery wikipedia , lookup
Cardiac contractility modulation wikipedia , lookup
ORIGINAL ARTICLE Liver function tests in patients with acute heart failure Jan Biegus1, Robert Zymliński1, Mateusz Sokolski1,2 , Sylwia Nawrocka1, Paweł Siwołowski1, Joanna Szachniewicz1, Ewa A. Jankowska1,2 , Waldemar Banasiak1, Piotr Ponikowski1,2 1 Centre for Heart Diseases, 4th Military Hospital, Wrocław, Poland 2 Department of Heart Diseases, Faculty of Health Science, Wroclaw Medical University, Wrocław, Poland Key words Abstract acute heart failure, liver dysfunction, liver function tests, prognosis Introduction Acute heart failure (AHF) is associated with multiorgan dysfunction, which may unfa‑ vorably affect prognosis. Objectives We investigated the prevalence, clinical determinants, and prognostic consequences of abnormal liver function tests (LFTs) in population with AHF. Patients and methods We conducted a retrospective analysis of patients with AHF, in whom the fol‑ lowing LFTs were performed on admission: serum bilirubin, aspartate transaminase (AST), alanine transaminase (ALT), and albumin. Abnormal LFTs were defined as the elevation above the upper normal limit of bilirubin, AST, and ALT, or reduction below the lower normal limit of albumin. Results The study involved 189 patients (age, 68 ±11 years; men, 68%; de novo AHF, 25%). On admis‑ sion, abnormal LFTs were observed in 46% of the patients for AST, 31% for ALT, 33% for bilirubin, and 44% for albumin. Only 29% of the patients had all LFTs within the normal ranges. The following variables were independently related to abnormal LFTs: high hemoglobin and N‑terminal pro‑B‑type natriuretic peptide (NT‑proBNP) levels for AST; high hemoglobin, bilirubin, and NT‑proBNP levels for ALT; high hemoglobin, low sodium levels, and dilated right ventricle for bilirubin; and high NT‑proBNP levels for albumin (all P <0.05). In 21 patients, hemodynamic monitoring was performed, which revealed that among LFTs only elevated bilirubin independently correlated with higher right atrial pressure (P <0.005). In a univariate Cox model, among LFTs, low albumin and markedly elevated AST and ALT (>3 times above the upper normal limit) were associated with increased mortality during 180‑day follow‑up. Conclusions Abnormal LFTs are common in patients with AHF and may have prognostic relevance. Among them, only elevated bilirubin was correlated with impaired hemodynamic parameters. Correspondence to: Jan Biegus, MD, Ośrodek Chorób Serca, 4. Wojskowy Szpital Kliniczny, ul. Weigla 5, 50-981 Wrocław, Poland, phone/fax: +48‑71-766‑02‑50, e‑mail: [email protected] Received: July 25, 2012. Revision accepted: October 3, 2012. Published online: October 4, 2012. Conflict of interest: none declared. Pol Arch Med Wewn. 2012; 122 (10): 471-479 Copyright by Medycyna Praktyczna, Kraków 2012 Introduction The pathophysiology of acute heart failure (AHF) is complex and not fully un‑ derstood.1 Only recently, it has been hypothesized that an episode of acute decompensation affects not only the cardiovascular system itself, but may also deteriorate the function of other organs, caus‑ ing serious clinical consequences. Traditionally, this phenomenon has been investigated on the cardio‑ renal axis and the term “cardiorenal syndrome” has been proposed to link cardiac dysfunction with the subsequent deterioration in renal function.2 The interactions between impaired hemodynamics, characterizing the early stage of AHF, with other key organs, have been rather poorly described. Liver function abnormalities have been well characterized mainly in patients with chronic heart failure (CHF).3‑8 The available data show that organ hypoperfusion as a consequence of low cardiac output and tissue congestion may lead to liver dysfunction.9 Additionally, sever‑ al reports have linked abnormal liver function tests (LFTs) with poor outcome in the chron‑ ic setting.3,6-8 Surprisingly, there are only few reports comprehensively assessing liver func‑ tion in AHF.10 Thus, we designed a study to detect the prevalence of liver dysfunction in well‑char‑ acterized, typical, contemporary population of patients admitted to the hospital due to AHF. ORIGINAL ARTICLE Liver function tests in patients with acute heart failure 471 Table 1 Baseline clinical and laboratory characteristics of patients with acute heart failure Parameter male sex 129 (68) age, y 68 ±11 SBP, mmHg 127 ±31 LVEF, % 35 ±14 LVEDD, mm 61 ±10 de novo AHF 47 (25) etiology of HF ischemic hypertension other 107 (57) 35 (18) 47 (25) comorbidities myocardial infarction hypertension atrial fibrillation diabetes 64 (34) 110 (58) 90 (48) 77 (41) medications (prior to hospitalization) ACEI/ARB β‑blockers aldosterone antagonists diuretics digoxin VKA/ASA 126 (67) 130 (69) 78 (41) 140 (74) 42 (22) 120 (63) liver function tests AST, IU/l ALT, IU/l bilirubin, mg/dl albumin, mg/dl 29 (21–45) 25 (17–47) 1.2 (0.8–1.9) 3.8 (3.5–4.1) hemoglobin, g/dl 13 ±1.8 sodium, mmol/l 139 ±4 creatinine, mg/dl 1.35 ±0.5 BUN, mg/dl 24 (19; 33) NT‑proBNP, pg/ml 5468 (3158–11846) troponin I, ng/ml 0.05 (0.03–0.16) Data are shown as number (%), mean ± SD or median (IQR). Abbreviations: ACEI – angiotensin‑converting enzyme inhibitor, AHF – acute heart failure, ALT – alanine transaminase, ARB – angiotensin receptor blocker, ASA – acetylsalicylic acid, AST – aspartate transaminase, BUN – blood urea nitrogen, HF – heart failure, IQR – interquartile range, LVEDD – left ventricular end‑diastolic diameter, LVEF – left ventricular ejection fraction, NT‑proBNP – N‑terminal pro‑B‑type natriuretic peptide, SBP – systolic blood pressure, SD – standard deviation, VKA – vitamin K antagonist We also aimed to establish potential determinants of abnormal LFTs and their impact on the pa‑ tient’s outcome. Patients and methods Study population We performed a retrospective analysis of all pa‑ tients who were hospitalized at the Centre of Heart Diseases, 4th Military Hospital, Wrocław, Poland, with primary diagnosis of AHF between January 2009 and October 2010, in whom the following LFTs were available on admis‑ sion: serum bilirubin, aspartate transaminase (AST), alanine transaminase (ALT), and albu‑ min. Patients with known liver disease or with clinical diagnosis of acute coronary syndrome on admission were excluded from the analysis. During recruitment, there were 242 hospital‑ ized patients with the diagnosis of AHF, of whom 189 (78%) had all LFTs available at baseline and this group constituted the study population. They 472 were predominantly men (68%), with the mean age of 68 ±11 years; 25% of the patients had de novo AHF and 57% had ischemic heart failure. The baseline clinical and laboratory characteristics of the study population are presented in TABLE 1 . Every patient underwent standard clinical eval‑ uation and transthoracic echocardiography with the measurements of the left ventricular ejec‑ tion fraction (LVEF, Simpson method) and the right and left ventricular end‑diastolic dimen‑ sions (RVEDD, LVEDD). Patients were treated in accordance with the recommendations of the European Society of Cardiology.11 The study protocol was approved by the local ethics committee, and the study was conducted in accordance with the Helsin‑ ki Declaration. Laboratory measurements On admission, the fol‑ lowing laboratory parameters were assessed in all patients using standard methods: 1) LFTs: AST, ALT, bilirubin, and albumin; in order to assess the prevalence of liver dysfunction we defined ab‑ normal LFTs as the values above the upper normal limit for AST, ALT, bilirubin (38 IU/l, 35 IU/l, and 1.3 mg/dl, respectively) or below the lower normal limit for albumin (3.8 mg/dl); additionally, we de‑ fined a marked elevation of AST and ALT as exceed‑ ing 3 times the upper normal limit; 2) blood count: hemoglobin, leukocytes, platelets; 3) renal function: creatinine and blood urea nitrogen (BUN); 4) elec‑ trolytes: sodium (NA+), potassium (K+); 4) periph‑ eral blood gases: pH, sO2, pCO2, serum osmolarity, lactate; 5) plasma N‑terminal pro‑B‑type natriuret‑ ic peptide (NT‑proBNP) by an immunoenzymatic method (Siemens, Marburg, Germany); 6) cardiac troponin I (TnI) by an immunoenzymatic method (single Dimension RxLMax, Siemens). Hemodynamic assessment In patients who re‑ quired hemodynamic monitoring on the basis of clinical evaluation, pulmonary artery catheter‑ ization using the Swan‑Ganz catheter was per‑ formed. Hemodynamic variables obtained during catheterization included: heart rate, blood pres‑ sure (systolic and diastolic), right atrial pressure (RAP), pulmonary artery pressure (PAP) (systol‑ ic, diastolic, mean), pulmonary capillary wedge pressure (PCWP), cardiac output (thermodilu‑ tion method). Each invasive assessment includ‑ ed from 3 to 5 separate injections of cold saline (10 ml), and the cardiac output was calculated as the mean value of all injections. The cardiac in‑ dex (CI) was determined as the cardiac output di‑ vided by the body surface area. Clinical follow‑up Data on survival was obtained directly from patients or their relatives (telephone contact), from the heart failure clinic database, or from the hospital system. No patient was lost to follow‑up. The primary endpoint was all‑cause death at 180 days. The length of follow‑up of sur‑ vivors and patients in whom the events occurred after 180 days were censored at 180 days. POLSKIE ARCHIWUM MEDYCYNY WEWNĘTRZNEJ 2012; 122 (10) % P <0.002 60 50 53 52 45 46 40 30 31 26 32 31 33 44 42 30 20 10 0 AST bilirubin total de novo AHF ALT albumin acutely decompensated Figure 1 Prevalence of abnormal liver function tests in acute heart failure Abbreviations: see table 1 Statistical analysis Continuous variables with normal distribution were described using means ± standard deviation; variables with skewed dis‑ tribution were described by medians with up‑ per and lower quartiles; categorized variables were presented as numbers and percentage. Vari‑ ables with skewed distribution were normalized by a logarithm. The statistical significance of dif‑ ferences between the groups were assessed us‑ ing t test, Mann‑Whitney U test, or χ2 test, where appropriate. Survival analyses were made using Cox propor‑ tional univariate hazard models, and the Kaplan‑ -Meier survival curves were demonstrated. The associations between clinical/laboratory variables (age, sex, etiology of heart failure, sys‑ tolic blood pressure on admission, hemoglobin, Na+, creatinine, BUN, LFTs, NT‑proBNP, pH, serum osmolarity, lactate, LVEF, LVEDD, and RVEDD) and abnormal LFTs were assessed by univariate logistic regression analyses. All vari‑ ables that were significantly associated with ab‑ normal LFTs in a univariate model were used to build a multivariate model. The relationships between variables were as‑ sessed using the Spearman rank coefficients. A P‑value below 0.05 was considered statistical‑ ly significant. Results Prevalence of abnormal liver function tests On admission, median [Q1; Q4] levels of AST, ALT, bilirubin, and albumin were 29 [21; 45] IU/l, 25 [17; 47] IU/l, 1.2 [0.8; 1.9] mg/dl, and 3.8 [3.5; 4.1] mg/dl, respectively. Abnormal LFTs were common in the study population, with the preva‑ lence of 46% for AST, 31% for ALT, 33% for bilirubin, and 44% for albumin (FIGURE 1). Only 54 patients (29%) had all LFTs within the normal range. The percentage of patients with elevated base‑ line bilirubin was higher in patients with dec‑ ompensated CHF compared with de novo AHF (52% vs. 26%, P <0.01) (FIGURE 1 ). We did not find any difference in the prevalence of other abnor‑ mal LFTs between decompensated CHF and de novo AHF. Comparison of patients with normal vs. abnormal liver tests Patients with abnormal transami‑ nases had higher hemoglobin, bilirubin, and NT‑proBNP levels and patients with elevated ALT had also lower albumin levels compared with pa‑ tients with normal transaminases (TABLE 2 ). Patients with elevated bilirubin had lower LVEF along with more dilated LVEDD and RVEDD (all P <0.05). This group had lower systolic blood pres‑ sure and Na+ concentration, and higher hemo‑ globin, BUN, AST, and NT‑proBNP levels com‑ pared with patients with normal bilirubin (all P <0.05) (TABLE 2 ). Compared with patients with normal albumin, patients with low albumin had ischemic etiology of heart failure less often, lower Na+, and higher NT‑proBNP concentration (all P <0.05) (TABLE 2 ). Associations of abnormal liver function tests with oth‑ er variables A multivariate regression model re‑ vealed that the following variables independent‑ ly predicted abnormal LFTs (TABLE 3 ): high hemo‑ globin and NT‑proBNP levels for AST; high hemo‑ globin, bilirubin, and NT‑proBNP levels for ALT; high hemoglobin, low sodium levels, and dilated RVEDD for bilirubin; and high NT‑proBNP levels for albumin (all P <0.05). Effect of liver dysfunction on mortality During 180‑day follow‑up, there were 40 deaths (21%). In a univari‑ ate Cox model, among abnormal LFTs, only low al‑ bumin and markedly elevated transaminases af‑ fected prognosis (hazard ratio [95% confidence in‑ terval]: for albumin, 2.13 [1.13–4.0]; for AST, 2.53 [1.06–6.03]; for ALT, 2.4 [1.06–5.44]; all P <0.05). Pa‑ tients with abnormal albumin as well as those with markedly elevated AST and ALT had significantly higher mortality compared with patients without these abnormalities: 29% vs. 15%, 39% vs. 19%, and 38% vs. 19%, respectively; all P <0.05. (FIGURE 2ABC). A multivariable Cox model of LFTs (adjusted for age, NT‑proBNP, and creatinine / estimated glomerular filtration rate [eGFR]) revealed that markedly ele‑ vated ALT and low albumin were significant prog‑ nosticators of adverse outcome (TABLE 4). ORIGINAL ARTICLE Liver function tests in patients with acute heart failure 473 Table 2 Comparison of clinical and laboratory characteristics in patients with acute heart failure with normal vs. abnormal liver function tests Parameter Normal AST, n = 130 Elevated AST, n = 59 Normal ALT, n = 126 male sex 90 (69) 39 (66) 88 (70) age, y 69 ±10 67 ±14 69 ±10 SBP, mmHg 129 ±32 123 ±28 129 ±34 LVEF, % 36 ±14 32 ±14 36 ±14 LVEDD, mm 61 ±10 63 ±11 61 ±10 de novo AHF 34 (26) 14 (24) 28 (22) etiology of HF ischemic hypertension other 77 (59) 27 (21) 25 (19) 30 (51) 8 (14) 23 (39) 74 (59) 26 (21) 25 (20) comorbidities myocardial infarction hypertension atrial fibrillation diabetes 49 (38) 82 (63) 57 (44) 56 (43) 15 (25) 28 (47) 33 (56) 41 (69) 49 (39) 77 (61) 58 (46) 57 (45) medications (before hospitalization) ACEI/ARB β‑blockers aldosterone antagonists diuretics digoxin VKA/ASA 84 (65) 88 (68) 51 (39) 96 (74) 29 (22) 82 (63) 42 (71) 42 (71) 27 (46) 44 (75) 13 (22) 38 (64) 88 (70) 89 (71) 54 (43) 99 (79) 30 (24) 84 (67) liver function tests AST, IU/l ALT, IU/l bilirubin, mg/dl albumin, mg/dl 23 (19–29) 20 (15–28) 1 (0.8–1.7) 3.9 (3.6–4.1) 64 (48–113) c 73 (45–114) c 1.5 (1.1–2.5)c 3.8 (3.4–4) 23 (18–29) 19 (15–25) 1 (0.8–1.7) 3.9 (3.6–4.1) hemoglobin, g/dl 12.9 ±1.8 13.6 ±1.7b 12.9 ±1.9 sodium, mmol/l 139 ±4 139 ±5 139 ±4 creatinine, mg/dl 1.4 ±0.6 1.3 ±0.4 1.4 ±0.6 BUN, mg/dl 25 (19–33) 24 (18–32) 25 (19–33) NT‑proBNP, pg/ml 4951 (2927–9536) 7273 (3626–15,858) a 4763 (2785–9515) 0.05 (0.03– 0.16) 0.06 (0.02– 0.15) 0.05 (0.03– 0.2) 4.0 ±1 2.1 ±0.5 80 ±15 34 ±10 10 ±5 17 ±5 4.2 ±1 2.2 ±0.5 81 ±20 34 ±15 10 ±8 18 ±6 3.8 ±1 2.0 ±0.5 80 ±15 35 ±10 11 ±6 17 ±5 troponin I, ng/ml hemodynamic parameters (n=21) cardiac output, l/min cardiac index, l/min/m2 MAP, mmHg PAP, mmHg RAP, mmHg PCWP, mmHg Data are shown as number (%), mean ± SD, or median (IQR). a P <0.05, b P <0.01, c P <0.001 Abbreviations: MAP – mean arterial pressure, PAP – pulmonary artery pressure, PCWP – pulmonary capillary wedge pressure, RAP – right atrial pressure, others – see TABLE 1 Relationship between hemodynamic indices and liv‑ er function tests Based on clinical indications, 21 patients (11%) underwent hemodynamic as‑ sessment. There were no differences in the hemo‑ dynamic profile of patients with normal vs. ele‑ vated transaminases. Patients with elevated bili‑ rubin when compared with patients with normal bilirubin had lower CI (2.0 ±0.3 vs. 2.4 ±0.6 l/min), higher RAP (13 ±6 vs. 5 ±3 mmHg), and mean PAP (38 ±11 vs. 27 ±11 mmHg) (all P <0.05) (TABLE 4). Patients with hypoalbuminemia had lower mean arterial pressure compared with the remaining population (71 ±11 vs. 88 ±16 mmHg; P <0.05). 474 We found strong correlations between bilirubin and mean PAP (r = 0.49) and RAP (r = 0.59) (both P <0.05). A multivariable regression analysis revealed that RAP was independently associated with el‑ evated bilirubin (P <0.005) (FIGURE 3 ). There were no correlations between AST, ALT, albumin, and hemodynamic parameters. Discussion Several laboratory tests to asses liver function are available, each providing dif‑ ferent clinical information. Among them, AST and ALT, bilirubin, and albumin are often used. POLSKIE ARCHIWUM MEDYCYNY WEWNĘTRZNEJ 2012; 122 (10) Elevated ALT, n = 63 Normal bilirubin, n = 103 Elevated bilirubin, n = 86 Normal albumin, n = 105 Decreased albumin, n = 84 41 (65) 63 (61) 66 (77)a 70 (67) 59 (70) 68 ±12 70 ±10 67 ±12 69 ±11 68 ±11 125 ±26 134 ±33 120 ±26c 130 ±32 124 ±30 33 ±15 39 ±13 30 ±13c 36 ±13 34 ±15 63 ±10 59 ±11 65 ±8 61 ±10 62 ±10 20 (32) 35 (34) 12 (14)c 22 (21) 25 (30) 33 (52) 9 (14) 23 (37) 59 (57) 26 (25) 18 (17) 48 (56) 9 (10)b 29 (34)a 68 (65) 18 (17) 17 (19) 39 (46) 17 (20) 25 (34) 15 (24)a 33 (52) 32 (51) 20 (32) 36 (35) 69 (67) 47 (46) 44 (43) 28 (33) 41 (48)a 43 (50) 33 (38) 41 (42) 67 (68) 49 (50) 46 (47) 23 (28) 42 (52) 41 (51) 31 (38) 38 (60) 41 (65) 24 (38) 38 (60)b 12 (19) 36 (57) 63 (61) 62 (60) 27 (26) 65 (63) 13 (13) 60 (58) 63 (73) 68 (79)c 51 (59)c 75 (87)c 29 (34)c 60 (70) 78 (74) 75 (72) 46 (44) 79 (75) 20 (19) 70 (67) 48 (57) 55 (65) 32 (38) 61 (73) 22 (26) 50 (59) 58 (39–108) c 76 (47–108) c 1.4 (0.9–2.1)b 3.7 (3.5–4)a 26 (20 –37) 25 (17–43) 0.8 (0.6–1) 3.8 (3.5–4) 31 (21–57) a 26 (16–66) 2.0 (1.6–3)c 3.8 (3.5–4) 26 (20 –40) 25 (17–38) 1.1 (0.8–1.9) 4 (3.9–4.2) 29 (21–52) 26 (17–68) 1.3 (0.8–2.3) 3.5 (3.2–3.6)c 13.5 ±1.8b 12.8 ±1.9 13.5 ±1.8b 13.1 ±1.7 13.1 ±2 139 ±5 140 ±4 138 ±5 140 ±4 138 ±5b 1.3 ±0.4 1.4 ±0.6 1.3 ±0.4 1.27 ±0.6 1.4 ±0.5 c c 24 (18–32) 24 (17–31) 27 (21–35) 24 (19–31) 26 (16–35) 7858 (4160 –15,632) c 4782 (2707–11,431) 6488 (4159–12,504) a 4507 (2818–9363) 7978 (4159–15,043) b 0.06 (0.02– 0.14) 0.05 (0.03– 0.2) 0.5 (0.02– 0.2) 0.06 (0.03– 0.17) 0.05 (0.02– 0.14) 4.5 ±0.9 2.3 ±0.4 82 ±21 31 ±15 8 ±6 18 ±6 4.5 ±1.2 2.4 ±0.6 79 ±22 27 ±11 5 ±3 15 ±6 3.8 ±0.8 2.0 ±0.3a 82 ±14 38 ±11a 13 ±6a 19 ±5 3.9 ±0.7 2.1 ±0.4 88 ±16 37 ±10 10 ±6 19 ±4 4.3 ±1.1 2.2 ±0.6 71 ±11a 30 ±15 13 ±5 16 ±8 a Elevated transaminases are sensitive markers of liver injury, elevated bilirubin usually indicates cholestasis or extensive heme breakdown, and low albumin reflects impaired liver synthetic ca‑ pabilities. It is worth noting that the magnitude of transaminases’ elevation in different patho‑ logic conditions vary dramatically and the cut‑ offs for clinical significance are usually arbitrary. We assumed that AST and ALT exceeding 3 times the upper normal limit will define marked alter‑ ation of transaminases. Surprisingly, the data on the prevalence, pathophysiology, and clinical significance of abnormalities of each LFT in AHF remain limited.10,12 Our study shows that abnormal LFTs are com‑ mon in AHF, with 71% of the patients having at least 1 abnormal test on admission. To the best of our knowledge, this is the first report on the prevalence of abnormal LFTs in contempo‑ rary broad spectrum of patients with AHF. Pre‑ vious papers described this problem in select‑ ed subgroups of patients, e.g., the paper by Shi‑ nagawa et al.13 in decompensated patients with the ejection fraction below 40%.13 The most ORIGINAL ARTICLE Liver function tests in patients with acute heart failure 475 Table 3 Associations of abnormal liver function tests Abnormal liver test AST ALT bilirubin albumin Parameter Univariate model odds ratio (95% CI) Multivariate model odds ratio (95% CI) hemoglobin, g/dl 1.3 (1.08–1.56)b 1.3 (1.06–1.6)b bilirubin, mg/dl 1.3 (1.03–1.69)a NSd NT-proBNP, pg/ml 1.54 (1.1–2.15) 1.59 (1.1–2.28)a hemoglobin, g/dl 1.29 (1.07–1.55)b 1.25 (1.02–1.52)a bilirubin, mg/dl 4.0 (1.77–9.3) 2.5 (1.02–6.09)a NT-proBNP, pg/ml 1.46 (1.004–2.04)a 1.46 (1.01–2.12)a SBP, mmHg 0.98 (0.97–0.99) NSd hemoglobin, g/dl 1.25 (1.05–1.48) 1.29 (1.03–1.59)a Na+, mmol/l 0.87 (0.8–0.94)c 0.88 (0.8–0.96)b a c c b NT-proBNP, pg/ml b 1.48 (1.08–2.04) NSd LVEF, % 0.95 (0.92–0.97)c NSd LVEDD, mm 1.07 (1.03–1.1) NSd RVEDD, mm 1.1 (1.05–1.17)c serum osmolarity, mOsm c 1.07 (1.01–1.13)a c 0.92 (0.89–0.96) NSd NT-proBNP, pg/ml 1.58 (1.14–2.17) b 1.47 (1.06–2.06)a Na+, mmol/l 0.9 (0.84–0.97)b NSd a P <0.05, b P <0.01, c P <0.001, d P >0.05 Abbreviations: CI – confidence interval, NA – sodium, NS – nonsignificant, RVEDD – right ventricular end‑diastolic diameter, others – see TABLE 1 recent analysis of the EVEREST trial showed that the percentage of patients with abnormal LFTs varies from 17% for albumin to 62% for γ‑glutamyltranspeptidase.10 In CHF, the preva‑ lence of abnormal LFTs differs between the au‑ thors, ranging from 10% to 40%.3,5,9,10 Based on our data, we can conclude that the prevalence of LFT abnormalities in contemporary population with AHF is higher than that reported in clinical trials and that described in CHF patients. The lat‑ ter phenomenon may indicate that liver dysfunc‑ tion is a part of AHF pathophysiology. Our data also shows that patients with acutely decompen‑ sated CHF have higher prevalence of abnormal bilirubin, but not transaminases and albumin, when compared with de novo AHF. We can spec‑ ulate that the difference may reflect the different intensity of pathophysiological processes that lead to liver dysfunction in heart failure (acutely decompensated chronic compared with de novo). As the pathophysiology of the liver dysfunction in heart failure is not completely understood, one may speculate that some of the processes that lead to the development of cardiorenal syndrome may also influence bilateral cardiohepatic interactions.2 How‑ ever, our data do not support this hypothesis. Ab‑ normal LFTs were not associated with kidney func‑ tion tests; moreover, patients with abnormal LFTs were not characterized by worse kidney function when compared with patients with normal LFTs (only patients with elevated bilirubin had higher BUN but not creatinine). This can be explained by the fact that the liver, unlike the kidneys, has dual (portal vein and hepatic artery) blood supply, which should make this organ more resistant to decreased cardiac output and low perfusion state. 476 Somehow unexpectedly, we found that abnor‑ mal LFTs (all but albumin) were associated with high hemoglobin levels. It is surprising since ane‑ mia is a well‑established, independent factor of unfavorable outcome in heart failure.14,15 High hemoglobin may be a marker of low fluid status (hemoconcentration), which subsequently may lead to hypoperfusion and liver cell dysfunction/ damage. This pathomechanism may at least in part explain our findings. However, abnormal LFTs were also associated with elevated NT‑proBNP – a marker of fluid overload, which shows that pathophysiology of the phenomenon is more com‑ plex and needs to be further studied. Our data support the hypothesis that liver dys‑ function may have adverse effect on prognosis in AHF patients. Interestingly, unlike other authors, we did not observe higher mortality in patients with elevated bilirubin.10,13,16 It has already been demonstrated that in CHF bilirubin is indepen‑ dently associated with morbidity and mortali‑ ty.3 In a nontrial population of more than 16,500 CHF patients, Yu et al.5 observed that low albu‑ min and elevated bilirubin were risk factors of to‑ tal mortality. Batin et al.16 has also demonstrated that AST along with bilirubin have prognostic im‑ portance in CHF patients.16 Shinagawa et al.12,13 studied clinical implications of liver dysfunction in AHF and found elevated total bilirubin on ad‑ mission to be a marker of poor prognosis.12,13 Only recently, Ambrosy et al.10 identified high bilirubin and low albumin, but not elevated ALT, to have prognostic importance in AHF. The discrepancy with our findings may be explained by the fact that they analyzed the population of selected patients participating in the clinical trial. It may not accurately reflect liver dysfunction across the whole spectrum of AHF patients (the prev‑ alence of abnormal LFTs was lower than in our study). Additionally, they presented mortality analysis with longer follow‑up, and such prog‑ nosticators may be different than ours. We have demonstrated that patients with markedly elevated transaminases as well as those with low albumin concentration have significant‑ ly higher mortality at 180 days. It is worth noting that albumin concentration may not only reflect liver function itself but also may be a marker of hemodilution due to fluid overload and the pa‑ tient’s nutritional status. This is in line with our finding that low albumin was independently pre‑ dicted by elevated NT‑proBNP levels. With as few as 40 events at 180 days of follow‑ -up, we were restricted to perform multivariable analyses taking into account only 4 variables. Af‑ ter adjustment for age, NT‑proBNP (reflecting se‑ verity of AHF and being a well‑established prog‑ nosticator) and creatinine/eGFR (renal function), we found that markedly elevated ALT and low al‑ bumin remained significant prognosticators of ad‑ verse outcome. Replacing NT‑proBNP with LVEF in the model did not affect the results. Recent hemodynamic studies have demonstrat‑ ed that both elevated RAP and decreased CI results POLSKIE ARCHIWUM MEDYCYNY WEWNĘTRZNEJ 2012; 122 (10) 1.05 A 1.00 0.95 survival 0.90 0.85 0.80 0.75 0.70 0.65 0 20 40 60 80 100 120 140 160 180 200 140 160 180 200 140 160 180 200 time (days) B 1.0 survival 0.9 0.8 0.7 0.6 0.5 0 20 40 60 80 100 120 time (days) C 1.0 0.9 survival Figure 2 Kaplan‑Meier survival analysis comparing patients with: A – decreased albumin (black curve) vs. normal albumin (blue curve); B – markedly elevated AST (blue curve) vs. rest of the population (black curve); C – markedly elevated ALT (blue curve) vs. the remaining population (black curve) Abbreviations: see table 1 0.8 0.7 0.6 0.5 0 20 40 60 80 100 120 time (days) in the elevation of both transaminases and total bilirubin,9,12,17 whereas reduced CI alone leads to an increase in AST, ALT and direct but not total bilirubin.9 Thus, both hypoperfusion and conges‑ tion play an important role in the development of liver dysfunction in heart failure. Since most of these studies were performed in CHF, our data extend these findings to acute conditions. We ob‑ served a strong effect of deteriorated hemodynam‑ ics on bilirubin and a weak effect on other LFTs. Patients with elevated bilirubin had lower CI and higher pressures in the right side of the heart (RAP, PAP) and tended for higher PCWP. We observed a strong correlation between RAP and bilirubin, ORIGINAL ARTICLE Liver function tests in patients with acute heart failure 477 Table 4 Univariate and multivariable model of 180‑day all‑cause mortality Liver function tests Univariate model markedly elevated AST Multivariable modela hazard ratio (95% CI) P hazard ratio (95% CI) P 2.53 (1.06–6.03) <0.05 2.38 (0.98–5.75) 0.05 age, y 1.03 (1.00–1.07) <0.05 1.06 (0.99–1.06) 0.1 NT‑proBNP, pg/ml 1.0018 (1.0023–1.034) <0.05 1.0 (0.99–1.0) >0.05 creatinine, mg/dl 1.8 (1.24–2.65) <0.05 1.69 (1.09–2.59) <0.05 markedly elevated ALT 2.4 (1.06–5.44) <0.05 2.28 (1.01–5.16) <0.05 age, y 1.03 (1.00–1.07) <0.05 1.03 (0.99–1.06) 0.06 NT‑proBNP, pg/ml 1.0018 (1.0023–1.034) <0.05 1.0 (0.99–1.00) 0.5 creatinine, mg/dl 1.8 (1.24–2.65) <0.05 1.64 (1.06–2.54) <0.05 2.13 (1.13– 4.0) <0.05 2.19 (1.14–4.2) 0.05 age, y 1.03 (1.00–1.07) <0.05 1.03 (0.99–1.07) 0.051 NT‑proBNP, pg/ml 1.0018 (1.0023–1.034) <0.05 1.0 (0.99–1.00) 0.8 creatinine, mg/dl 1.8 (1.24–2.65) <0.05 1.74 (1.1–2.7) <0.05 low albumin Abbreviations: see TABLEs 1 and 3 References 2.0 r = 0.59009 1.8 1 Gheorghiade M, Pang PS. Acute heart failure syndromes. J Am Coll Cardiol. 2009; 53: 557-573. 2 Ronco C, Haapio M, House AH, et al. Cardiorenal syndrome. J Am Coll Cardiol. 2008; 52: 1527-1539. bilirubin (mg/dl) 1.6 1.2 3 Allen LA, Felker GM, Pocock S, et al. Liver function abnormalities and outcome in patients with chronic heart failure: data from the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) program. Eur J Heart Fail. 2009; 11: 170-177. 1.0 4 Felder L, Mund A, Parker JG. Liver function tests in chronic congestive heart failure. Circulation. 1950; 2: 286-297. 1.4 5 Yu SB, Cui HY, Qin M, et al. [The prevalence and prognostic value of liver function abnormalities in patients in patients with chronic systolic heart fail‑ ure]. Zhonghua Yi Xue Za Zhi. 2011; 18; 91: 2673-2637. Chinese. 0.8 0.6 6 Szyguła‑Jurkiewicz B, Wojnicz R, Lekston, et al. [Effect of elevated bili‑ rubin levels on the long‑term outcome in patients with chronic heart failure due to hypertension]. Pol Arch Med Wewn. 2007; 117: 227-233. Polish. 95% CI 0.4 0.2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 RAP (mmHg) Figure 3 Correlation between right atrial pressure and bilirubin in 21 patients subject to hemodynamic monitoring Abbreviations: see tables 2 and 3 478 indicating that bilirubin may be a marker of ve‑ nous pressure and tissue congestion. It is worth noting that since invasive hemodynamic mon‑ itoring is not a part of routine AHF patient as‑ sessment, it was performed only in 21 patients in whom it was judged as clinically mandatory. A limitation of our study is the small and selected subpopulation of patients who underwent right heart catheterization. In conclusion, abnormal LFTs are common in AHF and may identify patients with worse out‑ come. Elevated bilirubin is strongly correlated with RAP indicating that congestion is an impor‑ tant process leading to hyperbilirubinemia in AHF; however, further investigations are needed. 7 Poelzl G, Ess M, Mussner‑Seeber Ch, et al. Liver dysfunction in chron‑ ic heart failure: prevalence, characteristics and prognostic significance. Eur J Clin Invest. 2012; 42: 153-163. 8 Poelzl G, Eberl Ch, Achrainer H, et al. Prevalence and prognostic signifi‑ cance of elevated gamma‑glutamyltransferase in chronic heart failure. Circ Heart Fail. 2009; 2: 294-302. 9 van Deursen VM, Damman K, Hillege HL, et al. Abnormal liver function in relation to hemodynamic profile in heart failure patients. J Cardiac Fail. 2010; 16: 84-90. 10 Ambrosy AP, Vaduganathan M, Huffman MD, et al. Clinical course and predictive value of liver function tests in patients hospitalized for worsening heart failure with reduced ejection fraction: an analysis of the EVEREST trial. Eur J Heart Fail. 2012; 14: 302-311. 11 Dickstein K, Cohen‑Solal A, Filippatos G, et al.; ESC Committee for Practice Guidelines (CPG). ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J. 2008; 29: 2388-2442. 12 Shinagawa H, Inomata T, Koitabashi T, et al. Prognostic significance of increased serum bilirubin levels coincident with cardiac decompensation in chronic heart failure. Circ J. 2008; 72: 364-369. 13 Shinagawa H, Inomata T, Koitabashi T, et al. Increased serum bilirubin levels coincident with heart failure decompensation indicate the need for in‑ travenous inotropic agents. Int Heart J. 2007; 48: 195-204. 14 Ezekowitz JA, McAlister FA, Armstrong PW. Anemia is common in heart failure and is associated with poor outcomes insights from a cohort of 12 065 patients with new‑onset heart failure. Circulation. 2003; 107: 223-225. Acknowledgments This research was financially 15 Kaldara‑Papatheodorou EE, Terrovitis JV, Nanas JN. Anemia in heart failure. Pol Arch Med Wewn. 2010; 120: 354-360. supported by the National Science Centre (Po‑ land) grant No. 6543/B/T02/2011/40 (granted to E.A.J and P.P.). 16 Batin P, Wickens M, McEntegart D, et al. The importance of abnor‑ malities of LFTs in predicting mortality in chronic heart failure. Eur Heart J. 1995; 16: 1613-1618. 17 Kubo SH, Walter BA, John DH. Liver function abnormalities in chronic heart failure influence of systemic hemodynamics. Arch Intern Med. 1987; 147: 1227-1230. POLSKIE ARCHIWUM MEDYCYNY WEWNĘTRZNEJ 2012; 122 (10) ARTYKUŁ ORYGINALNY Testy czynnościowe wątroby u pacjentów z ostrą niewydolnością serca Jan Biegus1, Robert Zymliński1, Mateusz Sokolski1,2 , Sylwia Nawrocka1, Paweł Siwołowski1, Joanna Szachniewicz1, Ewa A. Jankowska1,2 , Waldemar Banasiak1, Piotr Ponikowski1,2 1 Ośrodek Chorób Serca, 4 Wojskowy Szpital Kliniczny, Wrocław 2 Katedra i Klinika Chorób Serca, Wydział Nauk o Zdrowiu, Uniwersytet Medyczny we Wrocławiu, Wrocław Słowa kluczowe Streszczenie ostra niewydolność serca, rokowanie, wyniki testów czynnościowych wątroby, zaburzenia czynności wątroby Wprowadzenie Ostra niewydolność serca (ONS) jest zespołem chorobowym związanym z niewydol‑ nością wielonarządową, co może niekorzystnie wpływać na rokowanie. Cele Badaliśmy częstość występowania, determinanty kliniczne i konsekwencje prognostyczne nie‑ prawidłowych wyników testów wątrobowych (TW) w populacji pacjentów z ONS. Pacjenci i metody Przeprowadziliśmy retrospektywą analizę danych pacjentów z ONS, u których przy przyjęciu do szpitala wykonano następujące TW: stężenie bilirubiny w surowicy, aktywność aminotrans‑ ferazy asparaginianowej (AST) i alaninowej (ALT) oraz stężenie albuminy. Nieprawidłowe wyniki TW zdefiniowano jako wartości przewyższające górną granicę normy w przypadku bilirubiny, AST i ALT lub jako wartości poniżej dolnej granicy normy w przypadku albuminy. Wyniki Do analizy włączono 189 pacjentów (wiek: 68 ±11 lat, mężczyźni: 69%, de novo ONS 25%). Odsetek nieprawidłowych TW wynosił: AST u 46% pacjentów, ALT u 31%, bilirubinę u 33% i albuminę u 44%. Tylko 29% pacjentów miało wyniki wszystkich TW w granicach normy. Następujące zmienne miały niezależny wpływ na nieprawidłowe wyniki TW: w przypadku AST – duże stężenie hemoglobiny i wysoki poziom N‑końcowego propeptydu natriuretycznego typu B (NT‑proBNP); w przypadku ALT – duże stężenie hemoglobiny i bilirubiny oraz wysoki poziom NT‑proBNP; w przypadku bilirubiny – duże stężenie hemoglobiny, małe stężenie sodu i powiększony wymiar prawej komory serca; w przypadku albumin – wysoki poziom NT‑proBNP (wszystkie p <0,05). U 21 chorych przeprowadzono monitorowanie hemodynamiczne, które ujawniło, że w obrębie TW jedynie zwiększone stężenie bilirubiny niezależnie koreluje z podwyższonym ciśnieniem w prawym przedsionku (p <0,005). W jednoczynnikowym modelu Coxa, wśród TW, małe stężenie albumin i znacznie zwiększone AST i ALT (>3 razy powyżej górnej granicy normy) były związane ze zwiększonym ryzykiem zgonu w ciągu 180‑dniowej obserwacji. Wnioski Nieprawidłowe wyniki TW często występują u pacjentów z ONS i mogą mieć znaczenie prognostyczne. Wśród nich jedynie zwiększone stężenie bilirubiny było skorelowane z zaburzeniami parametrów hemodynamicznych. Adres do korespondencji: lek. med. Jan Biegus, Ośrodek Chorób Serca, 4 Wojskowy Szpital Kliniczny, ul. Weigla 5, 50-981 Wrocław, tel./fax: 71-766‑02‑50, e‑mail: [email protected] Praca wpłynęła: 25.07.2012. Przyjęta do druku: 03.10.2012. Publikacja online: 04.10.2012. Nie zgłoszono sprzeczności interesów Pol Arch Med Wewn. 2012; 122 (10): 471-479 Copyright by Medycyna Praktyczna, Kraków 2012 ARTYKUŁ ORYGINALNY Testy czynnościowe wątroby u pacjentów z ostrą niewydolnością serca 479